Project Summary/Abstract Our long-term goal is to understand the interactions between elements in noncoding regions of vertebrate mRNAs, and their cognate binding proteins, and how they integrate signals from disparate stimuli to control translation. Transcript-selective translational control is mediated by interactions of RNA-binding proteins to sequence/structural elements in non-coding regions, most often the 5'- or 3'-untranslated region (UTR) of the target transcript. In addition to protein-RNA interactions, RNA-RNA interactions also regulate gene expression, e.g., riboswitches in the UTR of bacterial mRNAs contain proximate structural elements that undergo conform- ational change in response to specific metabolites, and control translation. Recent experiments in our labora- tory suggest that human vascular endothelial growth factor (VEGF)-A mRNA contains adjoining elements that function as a novel stimulus-dependent, protein-directed riboswitch that exists in two metastable conforma- tions: a translation-silencing and a translation-permissive conformer. The binary switch is controlled by inte- gration of two signals, interferon (IFN)- and hypoxia, that regulate the amount or activity of the binding factors. Upon cell stimulation by IFN-, Glu-Pro tRNA synthetase (EPRS) is released from its residence from the tRNA multisynthetase complex and joins the GAIT (IFN-Gamma-Activated Inhibitor of Translation) complex. EPRS binds a defined, 29-nt GAIT element in the VEGF-A mRNA 3'UTR, stabilizing the translation-silencing conformer and inhibiting translation. However, superimposition of hypoxia on IFN- stimulation increases the level of heterogeneous nuclear ribonucleoprotein (hnRNP) L that binds a CA-rich element directly upstream of the GAIT element, stabilizing the translation-permissive conformer and allowing VEGF-A expression. We pro- pose the following specific hypothesis: The myeloid cell integrates signals from IFN- and hypoxia by regulating the relative amounts of hnRNP L and GAIT complex, which in turn dictate the conformation of the VEGF-A 3'UTR to either permit or suppress VEGF-A mRNA translation. We will test this hypothesis by pursuing the following Specific Aims: Aim 1: Determine sequences and secondary structures in the VEGF-A mRNA 3'UTR required for binary switch function. Aim 2: Determine the role of VEGF-A 3'UTR binding proteins in switch function. Aim 3: Investigate regulation of the VEGF-A 3'UTR binary switch by IFN- and hypoxia. We hypothesize that the switch evolved to maintain VEGF-A expression and angiogenesis in hypoxic, inflamma- tory tissues. Tumors, also residing in hypoxic, inflammatory sites, may take advantage of the VEGF-A switch to stimulate inward blood vessel growth to provide nourishment and permit tumor growth. Thus, the VEGF-A switch represents a novel therapeutic target to specifically inhibit tumor macrophage expression of VEGF-A. We also speculate that the VEGF-A switch may represent the founding member of a family of protein-directed riboswitches in vertebrates that integrate other physiological or pathological stimuli to control gene expression. Relevance Certain messenger RNAs respond to changes in their environment by altering their folding structure and their rate of expression of protein products. Although these riboswitches are found primarily in bacteria, we have found a similar switch in the mRNA encoding human vascular endothelial growth factor (VEGF), a protein critical for blood vessel formation. The VEGF riboswitch is sensitive to inflammation and hypoxia, two conditions found in the tumor environment, and an understanding of its mechanism may reveal insights into tumor growth and potential therapies to inhibit the process.